IOP Conference Series: Materials Science and Engineering

PAPER • OPEN ACCESS

Analytical Network Process (ANP) for priority setting of strategic roads

handling at Tebo Regency
To cite this article: Yosritzal et al 2019 IOP Conf. Ser.: Mater. Sci. Eng. 602 012096

View the article online for updates and enhancements.

This content was downloaded from IP address 158.46.142.223 on 06/09/2019 at 18:06

Conference on Innovation in Technology and Engineering Science IOP Publishing
IOP Conf. Series: Materials Science and Engineering 602 (2019) 012096 doi:10.1088/1757-899X/602/1/012096

Analytical Network Process (ANP) for priority setting

of strategic roads handling at Tebo Regency

Yosritzal1, J Permana2, B Istijono1, B Hidayat1, T Ophiyandri1, H Gunawan1

1
Lecturer, Civil Engineering Department, Andalas University, Padang, Indonesia
2Postgraduate Student, Civil Engineering Department, Andalas University, Indonesia

E-mail: yosritzal@eng.unand.ac.id

Abstract. Development of transportation infrastructure in the region aims to meet the

needs of the community to facilitate mobility of people, goods, and services. However,
the work of transportation infrastructure requires a large amount of cost while the
development budget is limited. Therefore, a careful determination of the priority of road
handling is urgently required. In this study, an analysis was conducted to determine
the priority of handling strategic roads in Tebo Regency. This study uses the Analytic
Network Process (ANP) method with 4 (four) criteria to determine the priority of road
handling, namely: engineering, economics, regional potential, and policies. It was
found that National Road - Pintas being the first priority with a weight of 0.252, then
the second priority for National Road - Lubuk Mandarsah with a weight of 0.23.The
third priority is Unit I Rimbo Bujang - Unit XI Rimbo Ulu Road with a weight of 0.211.

1. Introduction
In order to spur economic growth in the regions and central economic regions, the Tebo Regency
government has carried out various efforts. One of them is by improving transportation infrastructure
in the form of opening new roads, adding lengths of roads, widening roads, and road
rehabilitation/maintenance programs. The transportation infrastructure improvement program is not
easy for Tebo Regency as it requires a significant amount of money while the development budget is
limited. This has made it impossible for the Tebo Regency government to fulfill all of the road
construction needs in one budget year so that it must determine the priority of road handling
appropriately.
Previous research has introduced the Multi-Criteria Decision Making (MCDM) approach in
determining road handling priorities such as the Analytic Hierarchy Process (AHP) for example
[1][2][3]. The AHP method assumed that the criteria are independent, even though the
interdependence between criteria exists which could lead to a bias conclusion.
The Analytic Network Process (ANP) method is an approach which takes into account the
existence of dependencies in one group (inner dependence) and between different groups (outer
dependence) [4]. ANP method is able to overcome the shortcomings of AHP. In addition, the ANP
method is more complete and scientific than AHP, and the network structure in ANP is more
complicated than the hierarchical structure. Therefore, this research utilized the Analytic Network
Process (ANP) method in reviewing and analyzing the priority of strategic road handling in
TeboRegency.

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further
distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
Conference on Innovation in Technology and Engineering Science IOP Publishing
IOP Conf. Series: Materials Science and Engineering 602 (2019) 012096 doi:10.1088/1757-899X/602/1/012096

2. Literature Review

2.1. Analytic Network Process (ANP)

Analytic Network Process or ANPis the first mathematical theory that provides the possibility of
systematic dependency and feedback [5]. It is a new theory that extends the AHPwhich were both
introduced by Thomas L. Saaty [6].One of the advantages of this method is the way of obtaining
decisions and calculating the scale ratio. Priority as a scale of ratios is a fundamental element in
carrying out necessary arithmetic calculations by adding them on the same scale and multiplying them
with different scales that are meaningful to ANP [7].
Unlike the Analytic Hierarchy Process (AHP) method, ANP does not use hierarchies that are
linear but in the form of networks without having to set the level as used in AHP. The basic principle
of ANP is to include in the word of “influence” while in AHP is “a choice,” where in this case the use
of the word influence indeed requires more depth of meaning than the word choice [8].
Figure 1 shows that the linear hierarchy in AHP consists of the level of objectives, criteria, sub-
criteria, and alternatives, where each level has elements. Whereas in the ANP network model, the level
in AHP is called a cluster where it can contain criteria and alternatives called nodes.

Figure 1. Comparisons of hierarchy and network [4]

In the ANP network, alternatives have a dependency on criteria and also have interdependencies.
Likewise, the criteria in the ANP can be interdependent with other criteria as well as on alternatives.
The feedback is able to provide more accurate predictions than AHP because it can increase priorities
so that the results of ANP become more stable [9].

2.2. Ratter Agreement

Rater agreement is a measure that describes the level of agreement (approval) of the respondent (R 1-
Rn) on a problem in one cluster. One of the tools used to measure the rater agreement is Kendall's
Coefficient of Concordance (W; 0 <W ≤ 1).
To calculate Kendall's (W), for example, if there are raters (respondents) as much as m and the
subject as many as n will be given a rating from 1 to n, where rij is ranked on rater j against subject i,
then the following steps are taken [9]:

m
R r
i  ij (1)
j 1

2
Conference on Innovation in Technology and Engineering Science IOP Publishing
IOP Conf. Series: Materials Science and Engineering 602 (2019) 012096 doi:10.1088/1757-899X/602/1/012096

Calculate the average of the total ranking:

1
R  2 m n 1 (2)

Calculate the square of the deviation (S):

n

S  Ri  R 2 (3)
i1

So Kendall's (W) is obtained as:

12S
W
m 2 n 3  n (4)
The W value equal to 1 (W = 1) means that the respondents have perfect similarities in giving an
assessment or opinion. Whereas the W value equal to 0 or closer to 0indicates that the respondents
provide different responses [9].

2.3. Geometric Mean

When respondents more than one, an individual assessment of the respondents is needed, and then the
opinions of all respondents are grouped into one group using geometric mean calculations [4].
Questions in the form of pairwise comparison of all respondents will be combined to form a
consensus. Mean geometry is one of the average calculations that produce specific values with the
following formula [9]:

g  n ai  n a1 a2 an (5)

i1

3. Research methods

3.1. Determination of the studied Road

The road sections reviewed in this study are strategic regency roads which are the authority of the
regency government. This road segment is generally inter-city and connects cities in sub-regencies
with government administration centers, centers of economic activity, and other growth centers
including roads that connect between regency. The strategic road segment chosen is the road that has
been damaged and needs immediate handling. Based on some of these criteria and secondary data
obtained from the Office of Public Works and Spatial Planning of Tebo Regency, 5 (five) roads were
obtained with a total length of 61.4 km spread in 6 sub-regency in Tebo Regency as shown in Table 1.

3
Conference on Innovation in Technology and Engineering Science IOP Publishing
IOP Conf. Series: Materials Science and Engineering 602 (2019) 012096 doi:10.1088/1757-
899X/602/1/012096

Table 1. Strategic roads that need to be handled in Tebo Regency

No. No. Link Code Link Name Long (km)

1 025 Road 1 National Road – Pintas 11.9
2 026 Road 2 National Road – Paninjau 13.9
3 121 Road 3 National Road – Lubuk Mandarsah 18.5
4 067 Road 4 Block E Alai Ilir Road – Block C Alai Ilir 9.8
5 054 Road 5 The unit I Rimbo Bujang Road – Unit XI Rimbo Ulu 7.6
*Analysed from secondary sources

3.2. Data Collection

Data for this study was collected from interviews and questionnaires to respondents who are experts
and understand the road handling activities. The respondents consisted of 8 experts from the
TeboRegency government and 3 experts from the Tebo Regency House of Representative who were in
charge of infrastructure.Interviews and questionnaires were conducted by visiting respondents at their
offices, where the distribution and filling out of the questionnaire took approximately 2 months.

3.3. Identification of Criteria and Sub-Criteria

Based on the literature, the criteria and sub-criteria were identified and shown in Table 2. In the ANP
model, the criteria can also be referred to as a cluster while sub-criteria are defined as elements.

Table 2. Criteria and sub criteria for selection of alternative strategic roads.
Criteria/ Cluster Sub Criteria/ Element
A. Engineering a.1 Road Condition
a.2 Traffic Volume
B. Economy b.1 Investment Cost
b.2 Maintenance and Rehabilitation Cost
C. Regional Potential c.1 Agriculture
c.2 Trading and Services
c.3 Central Government
c.4 Education
c.5 Population Density
D. Policy d.1 Community Proposal
d.2 Government Program

3.4. Design of ANP Structure Model

Analytic Network Process (ANP) is the advancement of the Analytic Hierarchy Process (AHP) which
has a network structure model that is spread in all directions and involves cycles between groups and
linkages/dependencies on the same group. Determining the interrelationship/dependency relationship
was done by making a questionnaire with a voting method, based on research conducted by Kasirian
and Yusuff [10].
The questionnaire was given to three respondents, namely two (Head of Division or Kabid in the
TeboRegency Public Works and Spatial Planning Office and one in the Regional Development
Planning and Research Agency of TeboRegency. The number of respondents involved in determining
the relationship between these elements (N) is three people, so that if in a cell the number of
respondents who choose (Vij) is more or equal to Q, where Q is N/2 = 3/2 = 1.5, then the element that

4
Conference on Innovation in Technology and Engineering Science IOP Publishing
IOP Conf. Series: Materials Science and Engineering 602 (2019) 012096 doi:10.1088/1757-899X/602/1/012096

connects the cell is declared to have a relationship. Recapitulation of the results of the questionnaire
regarding the interconnected relationships between elements can be seen in Figure 2. Then based on
the dependency relationship between these elements, an ANP structure model was made for priority
handling of the regency strategic roads as shown in Figure 3.

Be Influenced
Criteria Engineering Econom y RegionalPotential Policy
RC TV IC M C A G TS C G ED PD C P G P
Econom y Engineering

RC 3 3 3 2 3 2 2 1 3 3

TV 3 3 3 0 2 1 1 0 3 3

IC 3 0 3 0 0 0 0 0 2 2

M C 3 0 3 0 0 0 0 0 2 2

A G 2 2 0 0 3 0 0 1 3 3
RegionalPotential

TS 3 0 0 0 0 0 2 3 3
Influence

C G 2 3 0 0 0 0 0 1 3 3
ED 2 3 0 0 0 0 0 1 3 3

PD 2 2 0 0 1 2 0 2 3 3

C P 1 1 3 3 0 0 0 0 0 1
Policy

G P 1 1 3 3 0 0 0 0 0 1

Figure 2. Recapitulation of questionnaire Figure 3. Alternative selection model of road

results regarding the relationship between section with ANP
relationships between elements.

4. Result and Discussion

4.1. Survey
The questionnaires were handed to respondents at their offices and taken back a few days later or by
agreement. Maximum allocated time for the respondents to complete the questionnaires was two
months starting from the day of the questionnaires were handed.
At the end of the allocated time, from 11 questionnaires distributed, 1 respondent failed to return
the questionnaire. Therefore, the number of returned questionnaires was 10. This was analyzed to
determine the priority road to be handled.

4.2. Data Analysis

Data were analyzed using the Analytic Network Process (ANP) method which was assisted by Super
Decisions 2.8.0 software. The data analysis stage was divided into 2. The first stage aims to process
the opinions and judgments of each respondent based on the weighting questionnaire they have filled
in to see the respondent's rater agreement on one problem in one cluster. In the second stage, the
consistency was tested, responses were put together using a geometric mean to obtain the results of
shared opinions and assessments which will become the final results for determining strategic road
handling regency in TeboRegency.

4.3. Priority Analysis for Cluster

Priority analysis for the cluster was based on rater agreement between individual and expert consensus
values or geometric mean values of 10 respondents. Figure 4 and Figure 5 show the results.

5
Conference on Innovation in Technology and Engineering Science IOP Publishing
IOP Conf. Series: Materials Science and Engineering 602 (2019) 012096 doi:10.1088/1757-899X/602/1/012096

Figure 4. Economic cluster priority. Figure 5. Policy cluster priority.

Figure 4 shows that the investment costs and maintenance and rehabilitation costs almost had the
same effect in determining which regency strategic roads were prioritized to be handled with weights
of 0.532 and 0.468, respectively. Respondents' opinion on the priority assessment of these two aspects
is very diverse which can be seen from the rater agreement which is quite low, that is equal to (W =
0.04). With the W value almost close to 0, it can be concluded that the number of respondents who
choose investment costs as a priority in the economic cluster is almost the same as respondents who
choose maintenance and rehabilitation costs.
In the policy cluster, as shown in Figure 5, the collective agreement of the experts revealed that
roadworks a program of the government are aspects that have a more significant influence in
determining the priority of regency strategic road handling with a weight of 0.557 compared to
roadworks derived from community proposals (weight is equal to 0.443. Even though the difference in
assessment given to these two aspects is not so significant, the agreement on the two aspects is quite
high, as indicated by the rater agreement value of (W = 0.36).

Figure 6. Engineering cluster priority. Figure 7. Regional potential cluster priority

Unlike the economic clusters and policy clusters where the two elements have similar priority
weights, the respondents' assessment of the elements in the engineering cluster has a different priority
weight as shown in Figure 6. The combined opinion of the respondents gives priority weight to aspects
of the road condition is 0.659, almost double compared to the priority weight given to the aspect of
traffic volume that is only 0.341. The assessment of the priorities of these two aspects is also
strengthened by the perfect rater agreement (W = 1). This means that all respondents agree that road
conditions are the most influential aspect in determining the priority of handling strategic roads when
viewed in terms of technique compared to aspects of traffic volume.
The next priority is a cluster of potential areas consisting of 5 elements. Figure 6 shows that the
aspects of trade and services and aspects of agriculture have a higher priority weight than other aspects
with almost the same value of 0.332 and 0.311 followed by aspects of population density with a
priority weight of 0.219. Whereas the other 2 aspects, namely the aspect of the central government and
aspects of education received very low voting. The weight is 0.07 and 0.068 respectively in the
potential regional cluster. The results of the assessment of the respondents were also supported by

6
Conference on Innovation in Technology and Engineering Science IOP Publishing
IOP Conf. Series: Materials Science and Engineering 602 (2019) 012096 doi:10.1088/1757-899X/602/1/012096

their high agreement on the weighting of each aspect in the potential cluster area with the rater
agreement (W = 0.856). The W value which is almost close to 1 indicates that the answers given by
the respondents are very homogeneous.

Figure 8. Handling priority of the strategic road.

The final goal of this study was to determine the priority sequence for handling the regency's
strategic road in Tebo Regency. From the results of the consensus of the opinions of all respondents
who were processed using the Analytic Network Process (ANP) method, priority was given to
handling strategic roads as shown in Figure 8.
Based on the weighting results in Figure 8, it can be seen that the experts chose Road 1 as the first
priority in the order of handling strategic roads with a weight of 0.252. Then The second order is road
3 with a weight of 0.231, then road 5 with a weight of 0.211 is in the third priority. Road 4 with a
weight of 0.202 becomes the fourth priority, while the last sequence with a weight that is much lower
than the other priority is 0.105 is the Road 2.
From the rater agreement value (W = 0.502), it can be seen that the agreement in giving priority to
strategic road handling by experts consisting of 10 respondents classified as high. This means that the
respondents did not differ in opinion on the determination of the strategic road handling sequence.
Based on the analysis above, the final priority sequence for handling the Regency Strategic Roads
in Tebo Regency as follows:
• Priority 1: Road1 (National Road – Pintas)
• Priority 2:Road 3 (National Road – Lubuk Mandarsah)
• Priority 3:Road 5 (Unit I Rimbo Bujang Road – Unit XI Rimbo Ulu)
• Priority 4:Road 4 (Blok E Alai Ilir Road – Blok C Alai Ilir)
• Priority 5:Road 2 (National Road – Paninjau)

5. Conclusion
In determining the priority handling of roads, the selection of criteria and sub-criteria dramatically
influences the final results. The selection of criteria and sub-criteria must consider many aspects and
be adapted to geographical conditions and socio-cultural conditions in the area to be studied. The
results of the analysis of the priority of each cluster show that there are four elements that are most
dominant in influencing the selection of regency strategic road handling, namely the agricultural
aspects, trade and service aspects and aspects of road conditions. This can be seen from the magnitude
of the priority given to the element compared to other elements in a cluster.
The final result of the priority sequence for handling the regency's strategic road in Tebo Regency
is the National Road - Pintas, National Road - Lubuk Mandarsah, Unit I Rimbo Bujang Road - Unit XI
Rimbo Ulu, Blok E Alai Ilir Road - Blok C Alai Ilir, and National Road - Paninjau. The respondent's
agreement in determining the priority order for handling the road is quite high as indicated by the rater
agreement value (W = 0.502).

7
Conference on Innovation in Technology and Engineering Science IOP Publishing
IOP Conf. Series: Materials Science and Engineering 602 (2019) 012096 doi:10.1088/1757-899X/602/1/012096

References
[1] Widyasari I 2015 Prioritas Penanganan Jalan di Kabupaten Bengkulu Utara (Priority of Road
Handling at North Bengkulu Regency) Jurnal Teknik Sipil, Magister Teknik Sipil Universitas
Sebelas Maret III [in Indonesian]
[2] Krismawati N E 2014 Analisis Prioritas Penanganan Ruas Jalan Strategies Untuk
Pengembangan Wilayah di Kabupaten Demak (Priority Setting of Strategic Roads Handling for
Regional Development at Demak Regency) Jurnal Wilayah dan Lingkungan 2 99-112 [in
Indonesian]
[3] Syawal A 2013 Perbandingan Skala Prioritas Penanganan Jalan di Kabupaten Bengkayang
antara Metode AHP dengan Metode Bina Marga (Comparison of Priority Road Handling Scale
at Bengkayang Regency between AHP Method and Highways Method) Jurnal Teknik Sipil
UNTAN 13 [in Indonesian]
[4] Saaty T L, Vargas L G 2006 Decision Making with The Analytic Network Process Economic,
Political, Social and Technological Applications with Benefits, Opportunities, Costs, and Risks
Pittsburgh: Springer
[5] Azis I J 2003 Analytic Network Process with Feedback Influence: a New Approach to impact
Study Paper Presented in Seminar Organized by the Department of Urban and Regional
Planning. University of Iullinois, Urbana-Campaign
[6] Rusydiana A S, Devi A 2013 Analytic Network Process, Pengantar Teori & Apliaksi (Analytic
Network Process, Introduction to Theories & Applications) Bogor: Smart Publishing
[7] Apriadi F, Findi M 2013 Solusi Peningkatan Sumberdaya Manusia Pada Baytul Maal wat
Tamwil (BMT) di Indonesia Melalui Pendekatan Analytic Network Process (ANP) (The
Solution of Human Resources Development for Baytul Maal wat Tamwil (BMT) in Indonesia:
an Analytic Network Process (ANP) Approach Jurnal al-Muzara’ah I
[8] Ascarya 2005 Analytic Network Process (ANP): Pendekatan Baru Studi Kualitatif (Analytic
Network Process (ANP): A New Approach to Qualitative Studies) Jakarta: Pusat Pendidikan dan
Studi Kebanksentralan, Bank Indonesia
[9] Ascarya (2011) The Persistence of Low Profit-and-Loss Sharing Financing in Islamic Banking:
The Case of Indonesia Jakarta: Center for Central Banking and Education Studies, Bank
Indonesia
[10] Kasirian M N, Yusuff R M 2009 Determining Interdependencies among Supplier Selection
European Journal of Scientific Research 35 76-84